danyashorokh

import matplotlib.pyplot as plt
from mpl_toolkits.axes_grid1 import ImageGrid

def plot_mosaic(images, titles, n_rows=2):
    
    fig = plt.figure(figsize=(16, 16.))
    grid = ImageGrid(fig, 111,
                     nrows_ncols=(n_rows, len(images) // n_rows),  # creates 2x2 grid of axes
                     axes_pad=0.3,  # pad between axes in inch.
                     )

danyashorokh

def binary_search_iterative(array, element):
    mid = 0
    start = 0
    end = len(array)
    step = 0

    while (start <= end):
        # print('Subarray in step {}: {}'.format(step, str(array[start:end+1])))
        step = step + 1
        mid = (start + end) // 2

danyashorokh

import matplotlib.pyplot as plt


def visualize(**images):
    'Plot images'
    n = len(images)
    plt.figure(figsize=(16, 5))
    for i, (name, image) in enumerate(images.items()):
        plt.subplot(1, n, i + 1)
        plt.xticks([])

danyashorokh

import tensorflow as tf


def conv2d_block(input_tensor, n_filters, kernel_size=3):
    x = input_tensor

    for i in range(2):
        x = tf.keras.layers.Conv2D(filters=n_filters, 
                                   kernel_size=(kernel_size, kernel_size))(x)
    

danyashorokh

from keras.preprocessing import image
from keras.applications.vgg16 import VGG16, preprocess_input as preprocess_input_vgg
from keras.applications.inception_v3 import InceptionV3, preprocess_input as preprocess_input_inception
from keras.applications.mobilenet_v2 import MobileNetV2, preprocess_input as preprocess_input_mobilenet
from keras.applications.resnet import ResNet50, preprocess_input as preprocess_input_resnet
from keras.models import Model, load_model
from keras.layers import Input, GlobalAveragePooling2D, GlobalMaxPooling2D
import numpy as np
import cv2

danyashorokh

# Detect hardware
try:
  tpu = tf.distribute.cluster_resolver.TPUClusterResolver() # TPU detection
except ValueError:
  tpu = None
  gpus = tf.config.experimental.list_logical_devices("GPU")
    
# Select appropriate distribution strategy
if tpu:
  tf.config.experimental_connect_to_cluster(tpu)

danyashorokh

import numpy as np

from keras.utils import Sequence
import imgaug as ia
import imgaug.augmenters as iaa
from imgaug.augmentables.segmaps import SegmentationMapsOnImage
ia.seed(1)

train_imgs = np.load('train_imgs.npy')
train_masks = np.load('train_masks.npy')

danyashorokh

import numpy as np

from keras.preprocessing.image import ImageDataGenerator
import imgaug as ia
import imgaug.augmenters as iaa
ia.seed(1)

train_imgs = np.load('train_imgs.npy')
train_masks = np.load('train_masks.npy')
val_imgs = np.load('val_imgs.npy')

danyashorokh

from keras.utils import multi_gpu_model
from keras import backend as K

print(K.tensorflow_backend._get_available_gpus())

n_gpus = 2

model = get_some_model()

gpu_model = multi_gpu_model(model, gpus=n_gpus)

danyashorokh

def mask_overlay(image, mask, color=(0, 255, 0)):
    """
    Helper function to visualize mask on the top of the object
    """
    mask = np.dstack((mask, mask, mask)) * np.array(color)
    mask = mask.astype(np.uint8)
    weighted_sum = cv2.addWeighted(mask, 0.5, image, 0.5, 0.)
    img = image.copy()
    ind = mask[:, :, 1] > 0    
    img[ind] = weighted_sum[ind]